RustBCA version 3.0 by drobnyjt · Pull Request #328 · lcpp-org/RustBCA

drobnyjt · 2026-08-10T16:23:25Z

This PR will bump RustBCA to v3.0.0. The increment in major version is due to different behavior caused by implementing a seeded RNG; however, all input files should be backwards-compatible and produce results within expected Monte Carlo error, with the exception of correctness fixes noted in the list below.

Correctness

Fixed Biersack-Varelas stopping power for ck!=1.0; split electronic stopping powers into their own functions and implemented testing of electronic stopping functions.
Added additional assertions on Ed, Eb values (must be >0)
Implemented recoil generation and particle advance routines using Duff et al.'s orthonormal basis instead of ad hoc switching between two cartesian bases - this should significantly increase the robustness and decrease the errors (when ux ~ 1) involved in the calculation of recoil locations and particle deflection. Additionally, this is slightly faster, as it requires fewer floating point operations.

Optimizations

Implemented LazyLock initialized Lookup Tables (LUTs) for functions of (Za, Zb); this offers speedups of up to ~30%, with some new constraints: Za, Zb must be integers, and Z_MAX is currently hardcoded at 120.
Changed all unnecessary invocations of powf to powi, sqrt, or cbrt. This change offered a ~5% speedup.
Removed some redundant function calls - ~5% speedup.

New features

New seeded PRNG; I've implemented a streaming PRNG, ChaCha8, that the user can seed manually. This makes RustBCA deterministic for all functions except some C/Fortran API functions where implementing it would be troublesome to the function calling.
New INTERPOLATEDPLUS electronic stopping mode - this mode allows mild adjustments of the Bragg peak magnitude with a new parameter, c_i, in BiersackVarelas stopping, S_BV = ((S_LS)^(-ci) + (S_BB)^(-ci))^(-1/ci).
added generalized RustBCA function wrappers; these allow one to call a complete RustBCA simulation from an input file dictionary, with the option to directly output lists of particles to memory or to print files as in the standalone code.

Other changes

deprecated HDF5 input
minor code cleanup/optimizations throughout
added many more integration tests, tightened epsilon on some pre-existing unit tests

… cbrt, etc.

…NG values; fixed simple_bca PRNG so that it gets a new rng with each particle.

…ers switched to thread-level seeds from prngs.

… algorithm for residue-preserving floating point addition

…duced non-determinism for multiple species.

…thread-local rng for starting values.

…a the LIBRUSTBCA_SEED environment variable. Added an option to Options to let the user set the PRNG seed in the standalone code; it defaults to 0.

… variable seeding with default 0 to all library functions.

…eedup (13%)

…values - is fp mult associative? Guess we will find out

…ther ~5-10% speedup...

Minor code fixes

* Updated pyo3 API to latest version. * Fixes and improvements to Python library using newest version of pyo3. * Cargo.toml updates. * Update non-physics packages. * Updates to non-physics packages required some minor code changes * Add inlining type hints

* Experimental triangular lookup tables for screening lengths and Coulomb constant * Added cfg to swap between 'reasonable' Z and full periodic table * Add tests for Z1 > Z2 to test_rustbca.py * Add Python function to compute screened coulomb scattering integrals and test * cleanup of lib.rs * Added scattering integral tests to workflow * Add different option test script to tests * Final version of triangular LUTs.

* Swapped out sin, cos with sin_cos where possible * Fixed bbe call - was calling average instead of actual, ignoring input option * Fixed incorrect bulk binding model in tests. * removed unnecessary invocations of screening_length * Put DEFAULTNEWTON back to 1e-6 * replace costly heap allocations in gauss-mehler with iterator solution * Don't need lazylock for GL points, can just make them consts * Removed further unnecessary heap allocations.

* Draft of generalized RustBCA python wrapper(s) * Added tests of rustbca_py to test_different_options.py * Implemented better python error handling * Add different interaction potentials to scattering_integrals() * used macros to streamline both lib.rs geometry types and main.rs geometry types * Attempt to fix module imports across lib, main

* Update to edition 2024. Fix no_mangle configuration flags to unsafe * Remove unnecessary nan-checking * Upgrade version number to 3.0 * Removed redundant screening length calls; wrapped into scattering integral calls. * Fixed tests. * add .gitignore * Remove HDF5 * Fix workflows.

* Add improved Python error handling * Add pyresult outputs to remaining python functions. * Make new get_seed function from env vars

drobnyjt · 2026-08-10T16:52:21Z

            -0.2912948318493851,
            -2.096765499656263,
-            19.16045452701010,
+            19.160_454_527_010_1,


I assume these literals are from clippy fix

Yep: https://help.klocwork.com/2026.1/en-us/reference/rs.clippy.lossy_float_literal.htm

Presumably the new written value in the highlighted case above (there are several others throughout the code) represents the actual float that the previous value would have represented.

drobnyjt and others added 30 commits June 23, 2026 10:21

Fix Biersack-Varelas stopping power.

c17d50d

Changed all invocations of powf(<integer>) to appropriate powi, sqrt,…

8c41b67

… cbrt, etc.

Fixed int->float

895d510

Fix to beta

83acc2a

Update rustbca_compile_check.yml

97e32ce

Remove redundant num_threads == 1 check

442a3f0

Draft of hard-coded seeded RNG

6d9f60c

Fixes to new rng for libRustBCA

1403c83

Fixed standalone code streaming index so chunks don't get repeated PR…

0f10616

…NG values; fixed simple_bca PRNG so that it gets a new rng with each particle.

Further fixes to seeding of lib.rs functions.

7e2783d

Two more fixes to prngs; one seeded moved outside a loop, and two oth…

4fea0ac

…ers switched to thread-level seeds from prngs.

Fix non-deterministic floating point reduce using Moller-Knuth TwoSum…

9799a39

… algorithm for residue-preserving floating point addition

Add citation to docstring for Moller-Knuth TwoSum

c2e1fd8

Update readme doctests to take advantage of seeded RNG determinism

95672a0

removed erroneous non-determinism in collision partner choice - intro…

e855c2a

…duced non-determinism for multiple species.

Removed another source of non-determinism - input.rs was calling the …

a5b2a7b

…thread-local rng for starting values.

specify num_threads in environment

fb477d0

specify branch in README.md

f7d7b47

Add additional check to examples in README to run with byexample.

30ab88c

Added optional seed.

71179fa

Added ability to set seed for Python functions with hardcoded seed vi…

8954307

…a the LIBRUSTBCA_SEED environment variable. Added an option to Options to let the user set the PRNG seed in the standalone code; it defaults to 0.

Added comment to README doctest

17deb45

Added a check on the 2D example in the doctests.

4a2fb66

Fix to options in distributions test.

0778770

Added -1 option to get seed from thread local PRNG; added environment…

afaf4c5

… variable seeding with default 0 to all library functions.

Missed a couple hardcoded 0 seeds; fixed doctests.

9e7a6b6

Add cfg statement to fix dead code warning in lib.rs

f8a0774

Remove unnecessary checkout from README.md

5ce3368

Add assertions on specific, deterministic values to test_rustbca.py

f3895fe

Fix to assertion in test_rustbca

3bb2b10

drobnyjt and others added 25 commits July 31, 2026 14:41

Cache in cargo.toml

1ed27e0

Trying a static lookup table for Lindhard screening lengths.

34ecb17

LUT ZBL as well; in the TiO2 example I got a small but significant sp…

5e9daf8

…eedup (13%)

Added some comments and minor cleanup

00e1d08

Further cleanup

d3c8777

First steps to make a LUT for LS stopping; this *might* break seeded …

8518652

…values - is fp mult associative? Guess we will find out

LS lookup table implemented.

c277a71

Added LUTs for LS stopping, BV mean ionization pot. Seems to gain ano…

1db6143

…ther ~5-10% speedup...

Remove reference to deprecated accelerated ions feature

2fa41ac

Remove reference to deprecated accelerated ions feature (2)

7de3c06

Switch from dynamic to static dispatch in quadrature

f4c4291

Switch deep recoil suppression to screening length lookup

c869745

Switch deep high energy ffps to screening length lookup

bfe7734

Replaced stray powfs in lib.rs

5e2c426

Merge pull request #318 from lcpp-org/minor_code_fixes

258dc22

Minor code fixes

LUT

20b3269

Remove limitation on max Z - 120 should be fine.

5dc7c3f

minor code cleaunup - typo, paste error, redundant ();

cf28058

Add improved Python error handling (#327)

d87907c

* Add improved Python error handling * Add pyresult outputs to remaining python functions. * Make new get_seed function from env vars

Merge branch 'main' into dev

70cf6f8

drobnyjt commented Aug 10, 2026

View reviewed changes

Updat sputtering_yield error prop. in lib.rs

5f91a61

drobnyjt merged commit 673bd3f into main Aug 11, 2026
4 checks passed

drobnyjt deleted the dev branch August 11, 2026 00:29

drobnyjt restored the dev branch August 11, 2026 00:30

Navigation Menu

RustBCA version 3.0 - #328

RustBCA version 3.0#328
drobnyjt merged 100 commits into
mainfrom
dev

drobnyjt commented Aug 10, 2026 •

edited

Loading

Uh oh!

drobnyjt Aug 10, 2026

Uh oh!

drobnyjt Aug 10, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

drobnyjt commented Aug 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Correctness

Optimizations

New features

Other changes

Uh oh!

drobnyjt Aug 10, 2026

Choose a reason for hiding this comment

Uh oh!

drobnyjt Aug 10, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

drobnyjt commented Aug 10, 2026 •

edited

Loading